Forward genetic screening in the zebrafish affords an unparalleled opportunity to discover genes required in human blood cell development, and whose alteration can lead to premalignant states or overt leukemia. This proposal tests two linked hypotheses: (i) genome-wide ethylnitrosourea (ENU) mutagenesis screens in the zebrafish can be used to identify dominant and recessive mutations that cause a deficiency or abnormal distribution of circulating granulocytes, implicating genes important in vertebrate myelopoiesis; and (ii) a subset of the genes discovered by this method will have human counterparts that contribute to myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), or to one of the congenital neutropenias that predispose to these malignancies. In preliminary studies, the zebrafish myeloperoxidase (zMpo) gene was cloned to be used in these screens as a granulocyte developmental marker, and its specificity for cells of the granulocytic lineage was demonstrated by RNA in situ analysis during development and adulthood in the fish. Also, detailed morphologic histochemical, electron microscopic and in situ analysis of cells in normal zebrafish blood and kidney (the hematopoietic organ of adult zebrafish) have been performed, and the results will serve as normal benchmarks for the analysis of myeloid cell development in mutant zebrafish lines recovered during screening. Mutant fish identified by in situ hybridization following ENU mutagenesis (Aim 1) will be analyzed to determine the cell developmental stage at which the mutation occurred (stem vs committed progenitor vs mature) (Aim 2). Next, the chromosomal location of each mutation will be mapped on the zebrafish genome, and examined for synteny with known regions of loss-ofheterozygosity (LOH) in human MDS/AML (Aim 3). Positional cloning (Aim 4) will focus on genes most likely to have deleted or mutated counterparts in these two neoplasias, both characterized by disordered granulocytic development. These zebrafish mutants may also have human homologues among the mutated genes contributing to recessive congenital blood diseases associated with granulocytopenia that can predispose to MDS/AML. Mutant zebrafish lines that harbor mutations in homologues of previously unidentified MDS/AML tumor suppressor genes will also serve as animal models to help identify genes in pathways leading to myeloid malignancy. A long-range goal is to use these models as a starting point for second-generation modifier screens to identify suppressors and enhancers of the genes causing myelopoietic defects, which may then be exploited as targets for therapeutic development. MDS and AML are currently extremely difficult to treat and are generally only curable with myeloablative therapy followed by hematopoietic stem cell transplantation.